Shockwave Surfers - team project

2022 NASA Space Apps Challenge - Tacoma Location

Selected challenge

"Can AI Preserve Our Science Legacy?" 🔗

The challenge is to develop a technique using Artificial Intelligence (AI) to improve the accessibility and discoverability of records in the public NASA Technical Report Server (NTRS).

Proposed solution 🔗

Improve the accessibility and discoverability of records in the NTRS

• The Search Algorithm (Librarian Analogy)

  1. Enter query through the api
  2. Receive a list of documents
  3. Drill down into the full text of the document
  4. Run NLP to get some statistics for ranking

• automatically read NTRS documents

  • summarize them (Pre-Trained),
  • generate text analytic data (Word Cloud, Frequency Distribution Visualisation, Question Generation and Answering),
  • produce a list of topic keywords to help researchers find the documents they need.
  • Delivering PDFs in a binder available for reading (with similar parts highlighted)

Reference

• Challenge specification pdf 🔗

• An example query 🔗 that returns json and defaults to the first 25 of 345 items.

• json viewers

  • online 🔗

  • jq tutorials 1, 2

    cd /data/infiles
    jq '.' query-result.json | less
    jq 'keys' query-result.json
    jq '.results | .[2]' query-result.json | less
    jq '.results | .[2] | .downloads' query-result.json | less
    jq '.results | .[2] | .downloads | .[0]' query-result.json
    jq '.results | .[2] | .downloads | .[0] | .links' query-result.json
    jq '.results | .[2] | .downloads | .[0] | .links | .fulltext' query-result.json
    

• python

  • json-module 🔗 (note there is a difference between load and loads !)